import numpy as np
import tensorflow.compat.v2 as tf
tf.enable_v2_behavior()
import pandas as pd
from tensorflow import keras
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import RobustScaler
from sklearn.preprocessing import MinMaxScaler
from matplotlib import pyplot
import plotly.graph_objects as go
import math
import seaborn as sns
from sklearn.metrics import mean_squared_error
np.random.seed(1)
tf.random.set_seed(1)
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, LSTM, GRU, Dropout, RepeatVector, TimeDistributed
from keras import backend
MODELFILENAME = 'MODELS/GRU_3d_TFM_2c'
TIME_STEPS=432 #3d
CMODEL = GRU
MODEL = "GRU"
UNITS=43
DROPOUT1=0.405
DROPOUT2=0.331
ACTIVATION='tanh'
OPTIMIZER='adadelta'
EPOCHS=56
BATCHSIZE=11
VALIDATIONSPLIT=0.1
# Code to read csv file into Colaboratory:
# from google.colab import files
# uploaded = files.upload()
# import io
# df = pd.read_csv(io.BytesIO(uploaded['SentDATA.csv']))
# Dataset is now stored in a Pandas Dataframe
df = pd.read_csv('../../data/dadesTFM.csv')
df.reset_index(inplace=True)
df['Time'] = pd.to_datetime(df['Time'])
df = df.set_index('Time')
columns = ['PM1','PM25','PM10','PM1ATM','PM25ATM','PM10ATM']
df1 = df.copy();
df1 = df1.rename(columns={"PM 1":"PM1","PM 2.5":"PM25","PM 10":"PM10","PM 1 ATM":"PM1ATM","PM 2.5 ATM":"PM25ATM","PM 10 ATM":"PM10ATM"})
df1['PM1'] = df['PM 1'].astype(np.float32)
df1['PM25'] = df['PM 2.5'].astype(np.float32)
df1['PM10'] = df['PM 10'].astype(np.float32)
df1['PM1ATM'] = df['PM 1 ATM'].astype(np.float32)
df1['PM25ATM'] = df['PM 2.5 ATM'].astype(np.float32)
df1['PM10ATM'] = df['PM 10 ATM'].astype(np.float32)
df2 = df1.copy()
train_size = int(len(df2) * 0.8)
test_size = len(df2) - train_size
train, test = df2.iloc[0:train_size], df2.iloc[train_size:len(df2)]
train.shape, test.shape
((3117, 7), (780, 7))
#Standardize the data
for col in columns:
scaler = StandardScaler()
train[col] = scaler.fit_transform(train[[col]])
<ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]]) <ipython-input-6-83cecdbc25f8>:4: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy train[col] = scaler.fit_transform(train[[col]])
def create_sequences(X, y, time_steps=TIME_STEPS):
Xs, ys = [], []
for i in range(len(X)-time_steps):
Xs.append(X.iloc[i:(i+time_steps)].values)
ys.append(y.iloc[i+time_steps])
return np.array(Xs), np.array(ys)
X_train, y_train = create_sequences(train[[columns[1]]], train[columns[1]])
#X_test, y_test = create_sequences(test[[columns[1]]], test[columns[1]])
print(f'X_train shape: {X_train.shape}')
print(f'y_train shape: {y_train.shape}')
X_train shape: (2685, 432, 1) y_train shape: (2685,)
#afegir nova mètrica
def rmse(y_true, y_pred):
return backend.sqrt(backend.mean(backend.square(y_pred - y_true), axis=-1))
model = Sequential()
model.add(CMODEL(units = UNITS, return_sequences=True, input_shape=(X_train.shape[1], X_train.shape[2])))
model.add(Dropout(rate=DROPOUT1))
model.add(CMODEL(units = UNITS, return_sequences=True))
model.add(Dropout(rate=DROPOUT2))
model.add(TimeDistributed(Dense(1,kernel_initializer='normal',activation=ACTIVATION)))
model.compile(optimizer=OPTIMIZER, loss='mae',metrics=['mse',rmse])
model.summary()
Model: "sequential" _________________________________________________________________ Layer (type) Output Shape Param # ================================================================= gru (GRU) (None, 432, 43) 5934 _________________________________________________________________ dropout (Dropout) (None, 432, 43) 0 _________________________________________________________________ gru_1 (GRU) (None, 432, 43) 11352 _________________________________________________________________ dropout_1 (Dropout) (None, 432, 43) 0 _________________________________________________________________ time_distributed (TimeDistri (None, 432, 1) 44 ================================================================= Total params: 17,330 Trainable params: 17,330 Non-trainable params: 0 _________________________________________________________________
history = model.fit(X_train, y_train, epochs=EPOCHS, batch_size=BATCHSIZE, validation_split=VALIDATIONSPLIT,
callbacks=[keras.callbacks.EarlyStopping(monitor='val_loss', patience=5, mode='min')], shuffle=False)
Epoch 1/56 220/220 [==============================] - 72s 327ms/step - loss: 0.7726 - mse: 0.9538 - rmse: 0.7752 - val_loss: 0.8796 - val_mse: 0.8136 - val_rmse: 0.8798 Epoch 2/56 220/220 [==============================] - 56s 254ms/step - loss: 0.7724 - mse: 0.9535 - rmse: 0.7751 - val_loss: 0.8783 - val_mse: 0.8113 - val_rmse: 0.8785 Epoch 3/56 220/220 [==============================] - 46s 210ms/step - loss: 0.7722 - mse: 0.9531 - rmse: 0.7749 - val_loss: 0.8770 - val_mse: 0.8089 - val_rmse: 0.8772 Epoch 4/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7720 - mse: 0.9529 - rmse: 0.7748 - val_loss: 0.8756 - val_mse: 0.8064 - val_rmse: 0.8758 Epoch 5/56 220/220 [==============================] - 46s 207ms/step - loss: 0.7718 - mse: 0.9526 - rmse: 0.7747 - val_loss: 0.8741 - val_mse: 0.8038 - val_rmse: 0.8744 Epoch 6/56 220/220 [==============================] - 46s 211ms/step - loss: 0.7717 - mse: 0.9525 - rmse: 0.7746 - val_loss: 0.8727 - val_mse: 0.8012 - val_rmse: 0.8729 Epoch 7/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7714 - mse: 0.9521 - rmse: 0.7744 - val_loss: 0.8711 - val_mse: 0.7985 - val_rmse: 0.8714 Epoch 8/56 220/220 [==============================] - 45s 207ms/step - loss: 0.7712 - mse: 0.9518 - rmse: 0.7743 - val_loss: 0.8696 - val_mse: 0.7957 - val_rmse: 0.8698 Epoch 9/56 220/220 [==============================] - 45s 206ms/step - loss: 0.7711 - mse: 0.9517 - rmse: 0.7742 - val_loss: 0.8681 - val_mse: 0.7930 - val_rmse: 0.8683 Epoch 10/56 220/220 [==============================] - 46s 207ms/step - loss: 0.7707 - mse: 0.9513 - rmse: 0.7739 - val_loss: 0.8665 - val_mse: 0.7902 - val_rmse: 0.8667 Epoch 11/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7706 - mse: 0.9512 - rmse: 0.7738 - val_loss: 0.8649 - val_mse: 0.7874 - val_rmse: 0.8652 Epoch 12/56 220/220 [==============================] - 46s 207ms/step - loss: 0.7704 - mse: 0.9510 - rmse: 0.7737 - val_loss: 0.8633 - val_mse: 0.7845 - val_rmse: 0.8636 Epoch 13/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7701 - mse: 0.9506 - rmse: 0.7735 - val_loss: 0.8617 - val_mse: 0.7817 - val_rmse: 0.8619 Epoch 14/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7699 - mse: 0.9504 - rmse: 0.7734 - val_loss: 0.8601 - val_mse: 0.7788 - val_rmse: 0.8603 Epoch 15/56 220/220 [==============================] - 45s 207ms/step - loss: 0.7697 - mse: 0.9501 - rmse: 0.7732 - val_loss: 0.8584 - val_mse: 0.7760 - val_rmse: 0.8587 Epoch 16/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7694 - mse: 0.9499 - rmse: 0.7730 - val_loss: 0.8568 - val_mse: 0.7731 - val_rmse: 0.8571 Epoch 17/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7693 - mse: 0.9498 - rmse: 0.7729 - val_loss: 0.8551 - val_mse: 0.7702 - val_rmse: 0.8554 Epoch 18/56 220/220 [==============================] - 45s 207ms/step - loss: 0.7690 - mse: 0.9494 - rmse: 0.7727 - val_loss: 0.8535 - val_mse: 0.7673 - val_rmse: 0.8538 Epoch 19/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7688 - mse: 0.9493 - rmse: 0.7726 - val_loss: 0.8518 - val_mse: 0.7644 - val_rmse: 0.8521 Epoch 20/56 220/220 [==============================] - 45s 207ms/step - loss: 0.7685 - mse: 0.9491 - rmse: 0.7723 - val_loss: 0.8502 - val_mse: 0.7615 - val_rmse: 0.8505 Epoch 21/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7684 - mse: 0.9489 - rmse: 0.7723 - val_loss: 0.8485 - val_mse: 0.7586 - val_rmse: 0.8488 Epoch 22/56 220/220 [==============================] - 45s 206ms/step - loss: 0.7682 - mse: 0.9487 - rmse: 0.7721 - val_loss: 0.8468 - val_mse: 0.7558 - val_rmse: 0.8472 Epoch 23/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7679 - mse: 0.9485 - rmse: 0.7720 - val_loss: 0.8452 - val_mse: 0.7529 - val_rmse: 0.8455 Epoch 24/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7677 - mse: 0.9484 - rmse: 0.7718 - val_loss: 0.8435 - val_mse: 0.7500 - val_rmse: 0.8439 Epoch 25/56 220/220 [==============================] - 46s 207ms/step - loss: 0.7674 - mse: 0.9480 - rmse: 0.7715 - val_loss: 0.8419 - val_mse: 0.7472 - val_rmse: 0.8422 Epoch 26/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7673 - mse: 0.9480 - rmse: 0.7715 - val_loss: 0.8402 - val_mse: 0.7443 - val_rmse: 0.8406 Epoch 27/56 220/220 [==============================] - 46s 210ms/step - loss: 0.7670 - mse: 0.9477 - rmse: 0.7713 - val_loss: 0.8385 - val_mse: 0.7414 - val_rmse: 0.8389 Epoch 28/56 220/220 [==============================] - 45s 206ms/step - loss: 0.7668 - mse: 0.9477 - rmse: 0.7712 - val_loss: 0.8369 - val_mse: 0.7386 - val_rmse: 0.8373 Epoch 29/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7666 - mse: 0.9475 - rmse: 0.7711 - val_loss: 0.8352 - val_mse: 0.7358 - val_rmse: 0.8356 Epoch 30/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7663 - mse: 0.9473 - rmse: 0.7708 - val_loss: 0.8336 - val_mse: 0.7330 - val_rmse: 0.8340 Epoch 31/56 220/220 [==============================] - 46s 207ms/step - loss: 0.7661 - mse: 0.9471 - rmse: 0.7707 - val_loss: 0.8319 - val_mse: 0.7301 - val_rmse: 0.8323 Epoch 32/56 220/220 [==============================] - 45s 207ms/step - loss: 0.7659 - mse: 0.9469 - rmse: 0.7706 - val_loss: 0.8302 - val_mse: 0.7273 - val_rmse: 0.8307 Epoch 33/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7657 - mse: 0.9468 - rmse: 0.7704 - val_loss: 0.8286 - val_mse: 0.7245 - val_rmse: 0.8290 Epoch 34/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7655 - mse: 0.9466 - rmse: 0.7703 - val_loss: 0.8269 - val_mse: 0.7217 - val_rmse: 0.8274 Epoch 35/56 220/220 [==============================] - 45s 206ms/step - loss: 0.7653 - mse: 0.9464 - rmse: 0.7701 - val_loss: 0.8253 - val_mse: 0.7189 - val_rmse: 0.8258 Epoch 36/56 220/220 [==============================] - 46s 208ms/step - loss: 0.7650 - mse: 0.9462 - rmse: 0.7699 - val_loss: 0.8236 - val_mse: 0.7162 - val_rmse: 0.8241 Epoch 37/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7648 - mse: 0.9462 - rmse: 0.7698 - val_loss: 0.8220 - val_mse: 0.7134 - val_rmse: 0.8225 Epoch 38/56 220/220 [==============================] - 45s 205ms/step - loss: 0.7646 - mse: 0.9459 - rmse: 0.7696 - val_loss: 0.8203 - val_mse: 0.7106 - val_rmse: 0.8209 Epoch 39/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7644 - mse: 0.9458 - rmse: 0.7695 - val_loss: 0.8187 - val_mse: 0.7079 - val_rmse: 0.8192 Epoch 40/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7642 - mse: 0.9457 - rmse: 0.7694 - val_loss: 0.8171 - val_mse: 0.7052 - val_rmse: 0.8176 Epoch 41/56 220/220 [==============================] - 42s 193ms/step - loss: 0.7640 - mse: 0.9456 - rmse: 0.7693 - val_loss: 0.8154 - val_mse: 0.7024 - val_rmse: 0.8160 Epoch 42/56 220/220 [==============================] - 44s 199ms/step - loss: 0.7639 - mse: 0.9456 - rmse: 0.7692 - val_loss: 0.8138 - val_mse: 0.6997 - val_rmse: 0.8144 Epoch 43/56 220/220 [==============================] - 44s 198ms/step - loss: 0.7636 - mse: 0.9452 - rmse: 0.7690 - val_loss: 0.8122 - val_mse: 0.6970 - val_rmse: 0.8127 Epoch 44/56 220/220 [==============================] - 43s 197ms/step - loss: 0.7634 - mse: 0.9453 - rmse: 0.7689 - val_loss: 0.8105 - val_mse: 0.6943 - val_rmse: 0.8111 Epoch 45/56 220/220 [==============================] - 45s 203ms/step - loss: 0.7633 - mse: 0.9453 - rmse: 0.7688 - val_loss: 0.8089 - val_mse: 0.6916 - val_rmse: 0.8095 Epoch 46/56 220/220 [==============================] - 45s 205ms/step - loss: 0.7630 - mse: 0.9450 - rmse: 0.7686 - val_loss: 0.8073 - val_mse: 0.6889 - val_rmse: 0.8079 Epoch 47/56 220/220 [==============================] - 46s 209ms/step - loss: 0.7628 - mse: 0.9449 - rmse: 0.7685 - val_loss: 0.8057 - val_mse: 0.6863 - val_rmse: 0.8063 Epoch 48/56 220/220 [==============================] - 46s 211ms/step - loss: 0.7626 - mse: 0.9448 - rmse: 0.7684 - val_loss: 0.8040 - val_mse: 0.6836 - val_rmse: 0.8047 Epoch 49/56 220/220 [==============================] - 49s 221ms/step - loss: 0.7624 - mse: 0.9446 - rmse: 0.7682 - val_loss: 0.8024 - val_mse: 0.6809 - val_rmse: 0.8031 Epoch 50/56 220/220 [==============================] - 49s 225ms/step - loss: 0.7622 - mse: 0.9446 - rmse: 0.7681 - val_loss: 0.8008 - val_mse: 0.6783 - val_rmse: 0.8014 Epoch 51/56 220/220 [==============================] - 63s 288ms/step - loss: 0.7621 - mse: 0.9445 - rmse: 0.7680 - val_loss: 0.7992 - val_mse: 0.6756 - val_rmse: 0.7998 Epoch 52/56 220/220 [==============================] - 100s 456ms/step - loss: 0.7618 - mse: 0.9444 - rmse: 0.7678 - val_loss: 0.7976 - val_mse: 0.6730 - val_rmse: 0.7982 Epoch 53/56 220/220 [==============================] - 69s 313ms/step - loss: 0.7617 - mse: 0.9444 - rmse: 0.7677 - val_loss: 0.7959 - val_mse: 0.6704 - val_rmse: 0.7966 Epoch 54/56 220/220 [==============================] - 60s 273ms/step - loss: 0.7614 - mse: 0.9443 - rmse: 0.7676 - val_loss: 0.7943 - val_mse: 0.6678 - val_rmse: 0.7950 Epoch 55/56 220/220 [==============================] - 102s 463ms/step - loss: 0.7613 - mse: 0.9444 - rmse: 0.7675 - val_loss: 0.7927 - val_mse: 0.6652 - val_rmse: 0.7934 Epoch 56/56 220/220 [==============================] - 84s 380ms/step - loss: 0.7611 - mse: 0.9442 - rmse: 0.7675 - val_loss: 0.7911 - val_mse: 0.6626 - val_rmse: 0.7918
import matplotlib.pyplot as plt
plt.plot(history.history['loss'], label='MAE Training loss')
plt.plot(history.history['val_loss'], label='MAE Validation loss')
plt.plot(history.history['mse'], label='MSE Training loss')
plt.plot(history.history['val_mse'], label='MSE Validation loss')
plt.plot(history.history['rmse'], label='RMSE Training loss')
plt.plot(history.history['val_rmse'], label='RMSE Validation loss')
plt.legend();
X_train_pred = model.predict(X_train, verbose=0)
train_mae_loss = np.mean(np.abs(X_train_pred - X_train), axis=1)
plt.hist(train_mae_loss, bins=50)
plt.xlabel('Train MAE loss')
plt.ylabel('Number of Samples');
def evaluate_prediction(predictions, actual, model_name):
errors = predictions - actual
mse = np.square(errors).mean()
rmse = np.sqrt(mse)
mae = np.abs(errors).mean()
print(model_name + ':')
print('Mean Absolute Error: {:.4f}'.format(mae))
print('Root Mean Square Error: {:.4f}'.format(rmse))
print('Mean Square Error: {:.4f}'.format(mse))
print('')
return mae,rmse,mse
mae,rmse,mse = evaluate_prediction(X_train_pred, X_train,MODEL)
GRU: Mean Absolute Error: 0.7311 Root Mean Square Error: 0.9390 Mean Square Error: 0.8817
model.save(MODELFILENAME+'.h5')
#càlcul del threshold de test
def calculate_threshold(X_test, X_test_pred):
distance = np.sqrt(np.mean(np.square(X_test_pred - X_test),axis=1))
"""Sorting the scores/diffs and using a 0.80 as cutoff value to pick the threshold"""
distance.sort();
cut_off = int(0.85 * len(distance));
threshold = distance[cut_off];
return threshold
for col in columns:
print ("####################### "+col +" ###########################")
#Standardize the test data
scaler = StandardScaler()
test_cpy = test.copy()
test[col] = scaler.fit_transform(test[[col]])
#creem seqüencia amb finestra temporal per les dades de test
X_test1, y_test1 = create_sequences(test[[col]], test[col])
print(f'Testing shape: {X_test1.shape}')
#evaluem el model
eval = model.evaluate(X_test1, y_test1)
print("evaluate: ",eval)
#predim el model
X_test1_pred = model.predict(X_test1, verbose=0)
evaluate_prediction(X_test1_pred, X_test1,MODEL)
#càlcul del mae_loss
test1_mae_loss = np.mean(np.abs(X_test1_pred - X_test1), axis=1)
test1_rmse_loss = np.sqrt(np.mean(np.square(X_test1_pred - X_test1),axis=1))
# reshaping test prediction
X_test1_predReshape = X_test1_pred.reshape((X_test1_pred.shape[0] * X_test1_pred.shape[1]), X_test1_pred.shape[2])
# reshaping test data
X_test1Reshape = X_test1.reshape((X_test1.shape[0] * X_test1.shape[1]), X_test1.shape[2])
threshold_test = calculate_threshold(X_test1Reshape,X_test1_predReshape)
test1_score_df = pd.DataFrame(test[TIME_STEPS:])
test1_score_df['loss'] = test1_rmse_loss.reshape((-1))
test1_score_df['threshold'] = threshold_test
test1_score_df['anomaly'] = test1_score_df['loss'] > test1_score_df['threshold']
test1_score_df[col] = test[TIME_STEPS:][col]
#gràfic test lost i threshold
fig = go.Figure()
fig.add_trace(go.Scatter(x=test1_score_df.index, y=test1_score_df['loss'], name='Test loss'))
fig.add_trace(go.Scatter(x=test1_score_df.index, y=test1_score_df['threshold'], name='Threshold'))
fig.update_layout(showlegend=True, title='Test loss vs. Threshold')
fig.show()
#Posem les anomalies en un array
anomalies1 = test1_score_df.loc[test1_score_df['anomaly'] == True]
anomalies1.shape
print('anomalies: ',anomalies1.shape); print();
#Gràfic dels punts i de les anomalíes amb els valors de dades transformades per verificar que la normalització que s'ha fet no distorssiona les dades
fig = go.Figure()
fig.add_trace(go.Scatter(x=test1_score_df.index, y=scaler.inverse_transform(test1_score_df[col]), name=col))
fig.add_trace(go.Scatter(x=anomalies1.index, y=scaler.inverse_transform(anomalies1[col]), mode='markers', name='Anomaly'))
fig.update_layout(showlegend=True, title='Detected anomalies')
fig.show()
print ("######################################################")
####################### PM1 ########################### Testing shape: (348, 432, 1) 1/11 [=>............................] - ETA: 0s - loss: 0.4818 - mse: 0.3149 - rmse: 0.4892
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy test[col] = scaler.fit_transform(test[[col]])
11/11 [==============================] - 1s 55ms/step - loss: 0.7119 - mse: 1.0029 - rmse: 0.7183 evaluate: [0.7119088768959045, 1.0029270648956299, 0.7183235287666321] GRU: Mean Absolute Error: 0.6089 Root Mean Square Error: 0.8515 Mean Square Error: 0.7251
anomalies: (113, 10)
###################################################### ####################### PM25 ########################### Testing shape: (348, 432, 1) 2/11 [====>.........................] - ETA: 0s - loss: 0.6671 - mse: 0.6902 - rmse: 0.6730
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
11/11 [==============================] - 1s 52ms/step - loss: 0.7523 - mse: 1.1187 - rmse: 0.7588 evaluate: [0.7523362636566162, 1.1186896562576294, 0.7587635517120361] GRU: Mean Absolute Error: 0.6400 Root Mean Square Error: 0.8695 Mean Square Error: 0.7560
anomalies: (7, 10)
###################################################### ####################### PM10 ########################### Testing shape: (348, 432, 1) 1/11 [=>............................] - ETA: 0s - loss: 0.5293 - mse: 0.4017 - rmse: 0.5366
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
11/11 [==============================] - 1s 60ms/step - loss: 0.7838 - mse: 1.1930 - rmse: 0.7902 evaluate: [0.7837892770767212, 1.1929787397384644, 0.7901870608329773] GRU: Mean Absolute Error: 0.6652 Root Mean Square Error: 0.8805 Mean Square Error: 0.7753
anomalies: (3, 10)
###################################################### ####################### PM1ATM ########################### Testing shape: (348, 432, 1) 1/11 [=>............................] - ETA: 0s - loss: 0.5530 - mse: 0.4300 - rmse: 0.5599
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
11/11 [==============================] - 1s 56ms/step - loss: 0.7826 - mse: 1.0412 - rmse: 0.7888 evaluate: [0.7825615406036377, 1.0412296056747437, 0.7887939810752869] GRU: Mean Absolute Error: 0.6730 Root Mean Square Error: 0.8721 Mean Square Error: 0.7606
anomalies: (0, 10)
###################################################### ####################### PM25ATM ########################### Testing shape: (348, 432, 1) 1/11 [=>............................] - ETA: 0s - loss: 0.5451 - mse: 0.4186 - rmse: 0.5520
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
11/11 [==============================] - 1s 60ms/step - loss: 0.7712 - mse: 1.0261 - rmse: 0.7774 evaluate: [0.7712005972862244, 1.0260939598083496, 0.7774139046669006] GRU: Mean Absolute Error: 0.6641 Root Mean Square Error: 0.8652 Mean Square Error: 0.7485
anomalies: (0, 10)
###################################################### ####################### PM10ATM ########################### Testing shape: (348, 432, 1) 1/11 [=>............................] - ETA: 0s - loss: 0.5213 - mse: 0.3960 - rmse: 0.5285
<ipython-input-17-e1f1d6df3b5c>:8: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
11/11 [==============================] - 1s 58ms/step - loss: 0.7643 - mse: 1.0415 - rmse: 0.7705 evaluate: [0.7642537951469421, 1.0415347814559937, 0.7704819440841675] GRU: Mean Absolute Error: 0.6565 Root Mean Square Error: 0.8664 Mean Square Error: 0.7506
anomalies: (0, 10)
######################################################